>> Live from the MIT campus in Cambridge, Massachusetts, it's theCUBE, covering the 12th annual MIT Chief Data Officer and Information Quality Symposium. Brought to you by SiliconANGLE Media. >> Welcome back to theCUBE's coverage of MIT's CDOIQ. I'm your host, Rebecca Knight, along with my cohost, Peter Burris. We're joined by Jeff McMillan. He is the managing director at Morgan Stanley. Well, thanks so much for coming on theCUBE, Jeff. >> Thanks for having me, it's great to be here. >> So you were just on a panel that was discussing the challenges and the opportunities of the CDO today. I mean, it is a mark of where the CDO role is, just by virtue of the fact that so many corporations are putting it front and center in their organizations. >> Yeah, I think what's interesting, though, is it is bit of a solution in search of a problem, and what I find the biggest challenge that many of these people are facing is that data in and of itself solves nothing, right? Unless you actually say, what business problem am I trying to solve, is it a risk problem, is it an efficiency play, is it a customer service issue, and then building your data solutions in support of that. Too many people start their journey by hiring 400 people, and they create data lineages and they have to create a dictionary and they put all these structures in place, and most of them fail, because they actually didn't figure out what they're solving for, and often times, very elegant and small solutions can actually drive a lot of positive outcomes, but the biggest mistake that, and we actually just discussed this on the panel, is knowing what you're solving for is the first step to be a successful chief data officer. >> Well, investments in infrastructure before outcomes fail no matter what they are, right? So whether it's an infrastructure of doing data analytics better, as you said, a whole bunch of clusters and a whole bunch of metadata management and other stuff, if it's not applied to some end, it's not going to get adopted. So we like to think we were talking in the opening thing, that one of the things that a chief data officer needs to do is acculturate the business to the idea of data being an asset, something that can be applied to work. And it's interesting in part because data can also help you choose what work you should apply it to. So talk a little bit about that. Does that resonate with you? >> I would totally agree with that, and it's not different, like when the first person created a business 2,000 years ago, somewhere along the line they said they needed somebody to keep track of the money, right? And the chief financial officer role sort of emerged, and then we had this thing where people actually came to work every day and they weren't really well trained and didn't understand their responsibilities, so we created the head of human resources. And I think these functions have evolved because as the business model grows, you need to have people to drive specific skills and competencies around these areas. And the truth is, in most organizations, we don't treat data like an asset. And part of it is the machinery, it's getting your Hadoop clusters up and putting your data meta and all that stuff. >> Or we confuse the assets of the technology with the assets that drive business value. >> That's right, and when people fail, it is rarely because they couldn't get the right data quality controls in place. They fail because they didn't get the right engagement model, and they didn't get the left hand and the right hand talking together, and at the core, data is not a data problem, it's an organizational problem. >> So there is this lack of consensus about where the CDO should sit, what his or her responsibilities mandate, scope, what do you think is the answer here? >> Well, we just got off the panel, and this was actually hotly debated, and there were two views on this that were highly divergent. >> But none of the other panelists are here today. >> Yeah, so my view's the right view. (laughter) Actually, I'll lay out both arguments. One of my colleagues on the panel was really driving this tech-focused approach, and her argument, which has some matter in fairness, is that so much of data is about the technology and the interplay and also the knowledge and the expertise and appreciation. You know, technology's been dealing with this problem for 25 years. No one was actually listening to them, right? So there is tremendous knowledge and expertise built up there. I took the other side of the equation, and I worked for the co-heads of our business, because it's not about the technology. And again, the challenges and the barriers to success are not technical in nature, it's leadership. And one thing that's interesting about data, and the reason that people have such a hard time with it, is that the problem and the solution to the problem often sit in two different cost centers. So getting somebody else to care about a problem that impacts you, when it actually doesn't drive your outcomes, is really hard, and that requires leadership and it requires collaboration. And sitting in a technology organization, by the way, I work with terrific technology folks, so this is not a disparagement on them, but sitting under the co-heads of the business, I am able to have those conversations with the other leaders of the business, and say listen, I know that you don't care about this, but for the best interest of the organization, we have to make these investments and let me explain, and those people think more holistically 'cause they're solving for the enterprise as opposed to their individual piece of technology. >> Which really is kind of you said, it requires leadership and it requires collaboration, but that also is one of the fundamental orientation of what great strategy should be. It's a way of cohering the mental model, getting everybody to agree on what the outcome and what the objective needs to be. >> Totally, and by the way, for those of us who are around in the late 90s. >> Not me! (laughter) >> When everyone hired the head of Internet strategy. This feels very much the same way, right? Everyone built websites and they had straight through processing and they sort of woke up a year and a half later, and they said, how has this gotten better? And they said oh, maybe we actually need to connect it to our infrastructure. >> I'll date myself. I remember when these conversations about whether or not we had a CIL, when we had a head of DP within HR, we had a head of DP within accounting, and there was whoa, what are we, the chief is responsible from my perspective, and I'd like to hear what you have to say, a chief anything is responsible for getting a return on the assets that are entrusted to them. >> Yeah, and that is 100% true. That being said, where you make your money is in the businesses, and I think to be really good at this job, you have to be very humble. And you can't make it about you and your goals and objectives, 'cause I have no goals and objectives outside of the goals and objectives of the business that I support. And part of what a lot of the challenge that people have, is they want to build empires, and I actually, I said to my boss, I have declared success when I'm an organization of one, because what I've been able to done is I've been able to set up the right controls, I've got the right people on the right jobs who understand, and the right technology, but the innovation is happening. It doesn't happen to my group. It happens away from my group. It happens when that 23-year-old who has got, with six weeks of visualization training, is sitting at 10 o'clock at night, figures out a better way to sell a municipal bond, because they spent 100 of their hours working on that. It's democratizing access to that, and it's really finding that right balance between control, ensuring the right data quality's in place, but also giving people the ability to innovate, and I think that's the perfect inflection point where you want to be. >> So what is the answer here? How would you give this remedy to other organizations in terms of the best practices that have emerged, and how to do this and do it right? >> Well, first and foremost, you got to know what your strategy is. I was on a panel with GE and General Motors. Their goals and objectives are very different than my goals and objectives. So don't leave this conference because Jeff McMillan did it this way at Morgan Stanley, and assume that that's the right answer for you. I think you have to first ask yourself, what are the most important objectives, and what is your strategy, 'cause the other thing I find is, you ask that question, a lot of businesses, even in this world in which, 2018, we talk about all the time, they don't have a clearly articulated strategy. And unless you have a strategy, putting data on the back end of that is not going to solve the problem. So first and foremost, you got to have a strategy. And then secondly, you got to put the right technical infrastructure in, there's a lot of plumbing that goes into this, and I'm going to gloss over it, but it's really important, and then you got to put the right organizational structure in place. I actually don't believe that you create a different parallel committee around this. The way we do it at our firm is we actually, the existing executive committee, is responsible for this, it's an additional function of them. We report into that function, and then you say, what is your business goals and objectives? Figure out where the gaps are, and then spend the time, money, and resources to solve and focus on that, and do it one problem at a time, and in doing that, you start to build this, what I'll describe as a data-centric or decision-centric culture. >> We call it data first, and so the way we tend to think about it, and I want to bounce this off of you is, you know, what's your business, what are the activities, the outcomes that are necessary to perform that business, what activities are necessary to achieve those outcomes, what data is necessary to perform those activities? >> That's right. >> Does that kind of follow? >> 100%, and also what processes, 'cause the other thing is that you talk to the data consultants, it's all about the data. And then you talk to the process consultants about the process, it's all about all of those things, and the point is that the data is the piece that sits, but there are many factors that influence that. Sometimes it's a data quality program. Sometimes it's a training program. Sometimes it's a technology issue. Sometimes it's a vendor supply issue. There's a whole host of reasons, and really the question is how do you use the data as the rallying point to say, this is the objective source of truth, and where is that objective source of truth, either not from a quality perspective, or from a business perspective, how does it impact those business, and always going back to that thing, 'cause there's truth in that attribute. >> And is that a culture issue? >> Well, it's a process of the technology, and it ultimately is a culture. And it's going back to the original comment, is do you see data as a problem, or do you see data as an opportunity? And I would argue, and I'm not going to speak for other companies, but in the world of finance, we live in bits, zeroes and ones, right? We are an information based business at the core, that happens to be delivering a financial services product. And in that world, that is our competitive advantage. I have a database of every single transaction that every client has ever given with us at Morgan Stanley. I know your risk tolerance level. I know where you live, I know whether you have children. That is a powerful source of knowledge, that if harnessed appropriately, allows to deliver a far, far superior solution to our clients and what they were getting previously. >> Great, well Jeff, thanks so much for coming on the show. It's really fun talking to you. >> Yeah, thank you so much. >> I'm Rebecca Knight. For Peter Burris, we will have more from MIT's CDOIQ coming up just after this.